The purpose of this project in the demography of aging is to analyze the mortality experience of Asian American elderly. Death rates, routinely constructed by combining vital statistics and census data, are much lower for the race category "Asian or Pacific Islander (API)" than those for any other race category - less than two-thirds of white rates, and are also lower than death rates in Japan, the country with highest life expectancy. Before concluding that this population is truly healthier, perhaps due to some combination of "healthy immigrant" effect or wealth, problems with data sources must be ruled out. Specifically, there is reason to suspect inconsistencies in race coding between death certificates and the census. Further, the aggregate API category may mask considerable heterogenity among the six major ethnic groups: Japanese, Chinese, Korean, Vietnamese, Asian Indian, and Filipino Americans. The proposed project would determine age- and sex-specific death rates for those 65+ in each of these six groups using methodology which avoids the pitfalls of combining vital status and census data by employing a single data source for numerators and denominators, the Master Beneficiary Record at the Social Security Administration (SSA). We would identify persons 65+ in each of the six groups using methods for Asian ethnic identification developed in a previous project, based on race code, given name, surname/maiden name and place of birth. SSA death records for identified persons will be compared with death certificate age and race data. We will address these research questions: Do age-specific, sex-specific death rates for the elderly differ among the six major Asian American Populations? Do death rates differ by nativity, i.e. U.S.-born versus foreign-born? How reliable are age and race information on death certificates for Asian American elderly? Does reliability vary by ethnic group, age, nativity, or state of death? What are the cause-specific death rates by ethnic group, and by nativity? This work will allow more accurate projections of longevity for these rapidly growing ethnic populations, foster a better understanding of the relationship between immigration and health, and will identify major causes of mortality in each group, thus informing future epidemiologic research.